Summary: Project 1 Stroke is the leading cause of serious adult disability in the United States. One of the most devastating impairments resulting from stroke is aphasia, a language impairment caused by left hemisphere damage involving cortical language areas. It is generally accepted that behavioral aphasia treatment is effective. Nevertheless, different patients experience very different degrees of benefit from aphasia treatment. Despite considerable differences in the response to aphasia treatment, the relationship between patient factors and treatment response is poorly understood and very few reliable prognostic indicators have been identified. This is a major problem, as both time and resources are wasted when clinicians do not know what patients are likely to respond to treatment, or which treatment best fits individual patients. The purpose of the current project is to develop a model that includes biographical and cognitive/linguistic factors to predict patients'response to aphasia treatment. Aphasia severity is one of the few factors that has been identified as a reliable predictor of performance in treatment; it is generally accepted that more severe aphasia is associated with poorer treatment outcomes. However, aphasia severity is a multidimensional construct and patients with similar overall severity scores might demonstrate very different language impairment profiles. To better understand how language impairment relates to treatment outcomes, the dual stream model (DS model;1) will be consulted. Specifically, we will test whether measures of proportional damage to the cortical areas that comprise the DS model improve prediction of aphasia treatment response, beyond biographical and cognitive/linguistic factors. Although the DS model is a functional model grounded in neuroanatomy, we expect measures of speech and language that assess processes supported by the two major components of the DS model ?the dorsal and ventral streams ? might be redundant with measures of cortical damage. To understand whether our predictive model can be generalized across different kinds of treatment foci, each patient will undergo treatment devoted to phonological stimulation and a separate treatment phase focusing on semantic stimulation. Ultimately, the goal here is to construct a predictive model that will be made available on- line so that clinicians can enter test scores from individual patients to predict how likely a given patient is to respond to treatment, as well as the focus of that treatment. There is a great need for prognostic indicators of aphasia treatment response. At the completion of our research, we will understand why some patients respond better to aphasia treatment than others. We have selected treatment approaches that are routinely used in clinical practice, allowing for immediate translation of the findings directly into patient management. The current project will yield a vast dataset that will be made publicly available allowing others to study further aphasia treatment response in relation to cognitive/linguistic and lesion factors.